/**
* JPASupport instances a and b are equals if either <strong>a == b</strong> or a and b have same
* </strong>{@link #key key} and class</strong>
*
* @param other
* @return true if equality condition above is verified
*/
@Override
public boolean equals(Object other) {
final Object key = this._key();
if (other == null) {
return false;
}
if (this == other) {
return true;
}
if (key == null) {
return false;
}
if (play.db.Model.class.isAssignableFrom(other.getClass()) && key.getClass().isArray()) {
Object otherKey = ((play.db.Model) other)._key();
if (otherKey.getClass().isArray()) {
return Arrays.deepEquals((Object[]) key, (Object[]) otherKey);
}
return false;
}
if (!this.getClass().isAssignableFrom(other.getClass())) {
return false;
}
return key.equals(((play.db.Model) other)._key());
}
@Override
public boolean equals(Object other) {
if (!(other instanceof SimilarityMatrix)) {
return false;
}
SimilarityMatrix otherMatrix = (SimilarityMatrix) other;
return otherMatrix.builtFrom.equals(builtFrom)
&& otherMatrix.xSubmissions.equals(xSubmissions)
&& otherMatrix.ySubmissions.equals(ySubmissions)
&& Arrays.deepEquals(otherMatrix.entries, entries);
}
/**
* Train a Deep Learning model, assumes that all members are populated If checkpoint == null,
* then start training a new model, otherwise continue from a checkpoint
*/
public final void buildModel() {
DeepLearningModel cp = null;
if (_parms._checkpoint == null) {
cp =
new DeepLearningModel(
dest(),
_parms,
new DeepLearningModel.DeepLearningModelOutput(DeepLearning.this),
_train,
_valid,
nclasses());
cp.model_info().initializeMembers();
} else {
final DeepLearningModel previous = DKV.getGet(_parms._checkpoint);
if (previous == null) throw new IllegalArgumentException("Checkpoint not found.");
Log.info("Resuming from checkpoint.");
_job.update(0, "Resuming from checkpoint");
if (isClassifier() != previous._output.isClassifier())
throw new H2OIllegalArgumentException(
"Response type must be the same as for the checkpointed model.");
if (isSupervised() != previous._output.isSupervised())
throw new H2OIllegalArgumentException(
"Model type must be the same as for the checkpointed model.");
// check the user-given arguments for consistency
DeepLearningParameters oldP =
previous._parms; // sanitized parameters for checkpointed model
DeepLearningParameters newP = _parms; // user-given parameters for restart
DeepLearningParameters oldP2 = (DeepLearningParameters) oldP.clone();
DeepLearningParameters newP2 = (DeepLearningParameters) newP.clone();
DeepLearningParameters.Sanity.modifyParms(
oldP, oldP2, nclasses()); // sanitize the user-given parameters
DeepLearningParameters.Sanity.modifyParms(
newP, newP2, nclasses()); // sanitize the user-given parameters
DeepLearningParameters.Sanity.checkpoint(oldP2, newP2);
DataInfo dinfo;
try {
// PUBDEV-2513: Adapt _train and _valid (in-place) to match the frames that were used for
// the previous model
// This can add or remove dummy columns (can happen if the dataset is sparse and datasets
// have different non-const columns)
for (String st : previous.adaptTestForTrain(_train, true, false)) Log.warn(st);
for (String st : previous.adaptTestForTrain(_valid, true, false)) Log.warn(st);
dinfo = makeDataInfo(_train, _valid, _parms, nclasses());
DKV.put(dinfo);
cp = new DeepLearningModel(dest(), _parms, previous, false, dinfo);
cp.write_lock(_job);
if (!Arrays.equals(cp._output._names, previous._output._names)) {
throw new H2OIllegalArgumentException(
"The columns of the training data must be the same as for the checkpointed model. Check ignored columns (or disable ignore_const_cols).");
}
if (!Arrays.deepEquals(cp._output._domains, previous._output._domains)) {
throw new H2OIllegalArgumentException(
"Categorical factor levels of the training data must be the same as for the checkpointed model.");
}
if (dinfo.fullN() != previous.model_info().data_info().fullN()) {
throw new H2OIllegalArgumentException(
"Total number of predictors is different than for the checkpointed model.");
}
if (_parms._epochs <= previous.epoch_counter) {
throw new H2OIllegalArgumentException(
"Total number of epochs must be larger than the number of epochs already trained for the checkpointed model ("
+ previous.epoch_counter
+ ").");
}
// these are the mutable parameters that are to be used by the model (stored in
// model_info._parms)
final DeepLearningParameters actualNewP =
cp.model_info()
.get_params(); // actually used parameters for model building (defaults filled in,
// etc.)
assert (actualNewP != previous.model_info().get_params());
assert (actualNewP != newP);
assert (actualNewP != oldP);
DeepLearningParameters.Sanity.update(actualNewP, newP, nclasses());
Log.info(
"Continuing training after "
+ String.format("%.3f", previous.epoch_counter)
+ " epochs from the checkpointed model.");
cp.update(_job);
} catch (H2OIllegalArgumentException ex) {
if (cp != null) {
cp.unlock(_job);
cp.delete();
cp = null;
}
throw ex;
} finally {
if (cp != null) cp.unlock(_job);
}
}
trainModel(cp);
// clean up, but don't delete weights and biases if user asked for export
List<Key> keep = new ArrayList<>();
try {
if (_parms._export_weights_and_biases
&& cp._output.weights != null
&& cp._output.biases != null) {
for (Key k : Arrays.asList(cp._output.weights)) {
keep.add(k);
for (Vec vk : ((Frame) DKV.getGet(k)).vecs()) {
keep.add(vk._key);
}
}
for (Key k : Arrays.asList(cp._output.biases)) {
keep.add(k);
for (Vec vk : ((Frame) DKV.getGet(k)).vecs()) {
keep.add(vk._key);
}
}
}
} finally {
Scope.exit(keep.toArray(new Key[keep.size()]));
}
}
@Override
Val apply(Env env, Env.StackHelp stk, AST asts[]) {
// Execute all args. Find a canonical frame; all Frames must look like this one.
// Each argument turns into either a Frame (whose rows are entirely
// inlined) or a scalar (which is replicated across as a single row).
Frame fr = null; // Canonical Frame; all frames have the same column count, types and names
int nchks = 0; // Total chunks
Val vals[] = new Val[asts.length]; // Computed AST results
for (int i = 1; i < asts.length; i++) {
vals[i] = stk.track(asts[i].exec(env));
if (vals[i].isFrame()) {
fr = vals[i].getFrame();
nchks += fr.anyVec().nChunks(); // Total chunks
} else nchks++; // One chunk per scalar
}
// No Frame, just a pile-o-scalars?
Vec zz = null; // The zero-length vec for the zero-frame frame
if (fr == null) { // Zero-length, 1-column, default name
fr = new Frame(new String[] {Frame.defaultColName(0)}, new Vec[] {zz = Vec.makeZero(0)});
if (asts.length == 1) return new ValFrame(fr);
}
// Verify all Frames are the same columns, names, and types. Domains can vary, and will be the
// union
final Frame frs[] = new Frame[asts.length]; // Input frame
final byte[] types = fr.types(); // Column types
final int ncols = fr.numCols();
final long[] espc = new long[nchks + 1]; // Compute a new layout!
int coffset = 0;
for (int i = 1; i < asts.length; i++) {
Val val = vals[i]; // Save values computed for pass 2
Frame fr0 =
val.isFrame()
? val.getFrame()
// Scalar: auto-expand into a 1-row frame
: stk.track(new Frame(fr._names, Vec.makeCons(val.getNum(), 1L, fr.numCols())));
// Check that all frames are compatible
if (fr.numCols() != fr0.numCols())
throw new IllegalArgumentException(
"rbind frames must have all the same columns, found "
+ fr.numCols()
+ " and "
+ fr0.numCols()
+ " columns.");
if (!Arrays.deepEquals(fr._names, fr0._names))
throw new IllegalArgumentException(
"rbind frames must have all the same column names, found "
+ Arrays.toString(fr._names)
+ " and "
+ Arrays.toString(fr0._names));
if (!Arrays.equals(types, fr0.types()))
throw new IllegalArgumentException(
"rbind frames must have all the same column types, found "
+ Arrays.toString(types)
+ " and "
+ Arrays.toString(fr0.types()));
frs[i] = fr0; // Save frame
// Roll up the ESPC row counts
long roffset = espc[coffset];
long[] espc2 = fr0.anyVec().espc();
for (int j = 1; j < espc2.length; j++) // Roll up the row counts
espc[coffset + j] = (roffset + espc2[j]);
coffset += espc2.length - 1; // Chunk offset
}
if (zz != null) zz.remove();
// build up the new domains for each vec
HashMap<String, Integer>[] dmap = new HashMap[types.length];
String[][] domains = new String[types.length][];
int[][][] cmaps = new int[types.length][][];
for (int k = 0; k < types.length; ++k) {
dmap[k] = new HashMap<>();
int c = 0;
byte t = types[k];
if (t == Vec.T_CAT) {
int[][] maps = new int[frs.length][];
for (int i = 1; i < frs.length; i++) {
maps[i] = new int[frs[i].vec(k).domain().length];
for (int j = 0; j < maps[i].length; j++) {
String s = frs[i].vec(k).domain()[j];
if (!dmap[k].containsKey(s)) dmap[k].put(s, maps[i][j] = c++);
else maps[i][j] = dmap[k].get(s);
}
}
cmaps[k] = maps;
} else {
cmaps[k] = new int[frs.length][];
}
domains[k] = c == 0 ? null : new String[c];
for (Map.Entry<String, Integer> e : dmap[k].entrySet()) domains[k][e.getValue()] = e.getKey();
}
// Now make Keys for the new Vecs
Key<Vec>[] keys = fr.anyVec().group().addVecs(fr.numCols());
Vec[] vecs = new Vec[fr.numCols()];
int rowLayout = Vec.ESPC.rowLayout(keys[0], espc);
for (int i = 0; i < vecs.length; i++)
vecs[i] = new Vec(keys[i], rowLayout, domains[i], types[i]);
// Do the row-binds column-by-column.
// Switch to F/J thread for continuations
ParallelRbinds t;
H2O.submitTask(t = new ParallelRbinds(frs, espc, vecs, cmaps)).join();
return new ValFrame(new Frame(fr.names(), t._vecs));
}
/**
* Verifies that the actual <code>Object</code> array is not equal to the given array. Array
* equality is checked by <code>{@link Arrays#deepEquals(Object[], Object[])}</code>.
*
* @param array the given array to compare the actual array to.
* @return this assertion object.
* @throws AssertionError if the actual <code>Object</code> array is equal to the given one.
*/
@Override
public ObjectArrayAssert isNotEqualTo(Object[] array) {
if (!Arrays.deepEquals(actual, array)) return this;
failIfCustomMessageIsSet();
throw failure(unexpectedEqual(actual, array));
}
public int ai_move(int[][] board) {
if (running) {
System.out.println(
"This AI appears to be running multiple ai_moves simultaneously. That can't be right.");
}
running = true;
try {
// System.out.println(2 + Math.random());
if (recording) {
if (out == null) {
try {
int ind = 1;
File f = null;
while (true) {
try {
Scanner sc = new Scanner(new File("AIReplay" + ind + ".txt"));
ind++;
} catch (Exception e) {
break;
}
}
out = new PrintWriter(new File("AIReplay" + ind + ".txt"));
filename = "AIReplay" + ind + ".txt";
out.println("AI Version: " + VERSION);
} catch (Exception e) {
System.out.println("Could not write to file.");
}
}
fprint(board);
}
// if (fml == null) fml = new PrintWriter (new File("fmldebug.txt"));
if (thisAIIsCheating && max(board) < 8) {
board[0][0] = GameGUI.win_target;
}
if (debug2) sc.nextLine();
if (debug2) print(board);
if (debug) System.out.println("New cycle.");
if (debug) sc.nextLine();
if (dumbai) name += "Dumby";
turn++;
if (!queue.isEmpty()) {
int temp = queue.removeFirst();
if (temp > 0) {
running = false;
return temp;
}
}
int boardsum = 0;
for (int i = 0; i < 4; i++) {
for (int j = 0; j < 4; j++) {
boardsum += board[i][j];
}
}
boolean report = debug;
/*if (Math.random() < 0.0001) {
report = true;
for (int i = 0; i < 4; i++) {
System.out.println(Arrays.toString(board[i]));
}
for (int i = 0; i < 4; i++) {
System.out.println(movable(board, i));
}
System.out.println();
sc.nextLine();
}*/
if (dumbai) {
if (!name.endsWith("Dumby")) name += "Dumby";
System.out.println(turn);
running = false;
if (turn % 600 == 599) return KeyEvent.VK_DOWN;
if (turn % 3 == 0) return KeyEvent.VK_UP;
if (turn % 6 < 3) return KeyEvent.VK_LEFT;
return KeyEvent.VK_RIGHT;
} else {
if (name.indexOf(".") < 0) name += VERSION;
}
// gamestart processing
/*if(board[0][0] == 0) {
if (board[1][0] > board[0][1]) {
return KeyEvent.VK_UP;
}
if (board[1][0] < board[0][1]) {
return KeyEvent.VK_LEFT;
}
if (Math.random() < 0.5) return KeyEvent.VK_UP;
return KeyEvent.VK_LEFT;
}*/
long[] pref = {10, 20, 1, 1}; // LEFT, UP, RIGHT, DOWN
// check if moving right/down is safe
boolean occupied = true;
for (int i = 0; i < 4; i++) {
if (board[0][i] == 0) occupied = false;
if (i < 3 && board[0][i] == board[0][i + 1]) occupied = false;
}
if (!occupied) {
// pref[2] -= 100000000;
}
occupied = true;
for (int i = 0; i < 4; i++) {
if (board[i][0] == 0) occupied = false;
if (i < 3 && board[i][0] == board[i + 1][0]) occupied = false;
}
if (!occupied) {
// pref[3] -= 100000000;
}
pref[0] += 5;
pref[1] += 5;
// System.out.println(6 + Math.random());
// simulate
sum_board = sum(board);
delta_sum_board_7over8 = delta(sum_board * 7 / 8);
if (debug) print(board);
max_depth = 0;
for (int m = 0; m < 4; m++) {
if (debug) System.out.println("Now testing move: " + m);
int[][] sim = simulate(board, m);
if (Arrays.deepEquals(sim, board)) {
if (out != null) out.println("Move " + m + " invalid; skipping");
if (GameGUI.out != null) GameGUI.out.println("Move " + m + " invalid; skipping");
continue;
}
long worst = (long) 1999999999 * 1000000000;
long avg = 0;
int numt = 0;
for (int i = 0; i < 4; i++) {
for (int j = 0; j < 4; j++) {
if (sim[i][j] > 0) continue;
sim[i][j] = 2;
long temp = predictor(sim, iter_max / (int) Math.pow((countBlank(sim) + 1), 1.6), 1);
if (temp < worst) worst = temp;
avg += 9 * temp;
sim[i][j] = 4;
temp = predictor(sim, iter_max / (int) Math.pow((countBlank(sim) + 1), 1.6), 1);
if (temp < worst) worst = temp;
avg += temp;
sim[i][j] = 0;
numt += 10;
}
}
if (countBlank(sim) == 0) {
long temp = predictor(sim, iter_max / (int) pow((countBlank(sim) + 1), 2), 1);
if (temp < worst) worst = temp;
avg += temp;
numt++;
}
avg /= numt;
worst = (worst_weight * worst + avg) / (worst_weight + 1);
if (countBlank(sim) >= 8 && max(board) < 64) worst = avg;
if (debug || debug2) System.out.println("Move " + m + " final eval: " + worst);
if (out != null) out.println("Move " + m + " final eval: " + worst);
if (GameGUI.out != null) GameGUI.out.println("Move " + m + " final eval: " + worst);
pref[m] += worst;
}
if (debug2) System.out.println("Max depth: " + max_depth);
if (out != null) out.println("Max depth: " + max_depth);
if (GameGUI.out != null) GameGUI.out.println("Max depth: " + max_depth);
// System.out.println(5 + Math.random());
// process output
int[] dir = new int[4];
dir[0] = KeyEvent.VK_LEFT;
dir[1] = KeyEvent.VK_UP;
dir[2] = KeyEvent.VK_RIGHT;
dir[3] = KeyEvent.VK_DOWN;
if (report) System.out.println("Pref: " + Arrays.toString(pref));
for (int i = 0; i < 4; i++) {
int best = 0;
for (int j = 0; j < 4; j++) {
if (pref[j] > pref[best]) {
best = j;
}
}
pref[best] = Long.MIN_VALUE;
if (movable(board, best)) {
if (report) {
report = false;
if (debug) System.out.println("Chosen: " + best);
if (debug) sc.nextLine();
}
// if (pref[best] < -50000000) queue.add(best - 2);
running = false;
return dir[best];
}
// System.out.println("Unmovable: " + best);
// System.out.println("Pref: " + Arrays.toString(pref));
}
System.out.println("???");
for (int i = 0; i < 4; i++) {
System.out.println(Arrays.toString(board[i]));
}
// sc.nextLine();
} catch (Exception e) {
e.printStackTrace();
}
running = false;
return KeyEvent.VK_LEFT;
}
public long predictor(int[][] board, int iters, int depth) { // returns future value, kinda
// debnum+=4;
// System.out.println(Math.random());
if (depth > max_depth) max_depth = depth;
// if (max(board) < 64 && depth == 1) return grade4(board);
int div = 0;
for (int i = 0; i < 4; i++) {
int[][] sim = simulate(board, i);
if (Arrays.deepEquals(sim, board)) continue;
div += countBlank(sim);
}
if (!movable(board, 0) && !movable(board, 1) && !movable(board, 2) && !movable(board, 3)) {
// if (max(board) == GameGUI.win_target) return grade(board);
return (long) -1999999999 * 3 * sum(board);
}
div *= 2;
if (div > iters) {
// debnum-=4;
return grade4(board);
}
iters /= div;
long best = (long) -1999999999 * 800000000;
if (debug) print(board);
for (int m = 0; m < 4; m++) {
// debnum--;
int[][] sim = simulate(board, m);
if (Arrays.deepEquals(sim, board)) continue;
if (debug) System.out.println("Simulating: " + m);
long worst = (long) 1999999999 * 800000000;
long avg = 0;
int numt = 0;
for (int i = 0; i < 4; i++) {
for (int j = 0; j < 4; j++) {
if (sim[i][j] > 0) continue;
sim[i][j] = 2;
long temp = predictor(sim, iters, depth + 1);
if (temp < worst) worst = temp;
avg += 9 * temp;
sim[i][j] = 4;
temp = predictor(sim, iters, depth + 1);
if (temp < worst) worst = temp;
avg += temp;
sim[i][j] = 0;
numt += 10;
}
}
if (countBlank(sim) == 0) {
// System.out.println("??");
long temp = predictor(sim, iter_max / (int) pow((countBlank(sim) + 1), 2), depth + 1);
if (temp < worst) worst = temp;
avg += temp;
numt++;
}
// avg -= worst;
avg /= numt;
if (debug) System.out.println("Result: " + worst);
if ((avg + worst_weight * worst) / (worst_weight + 1) > best)
best = (worst_weight * worst + avg) / (worst_weight + 1);
// if (worst > best) best = worst;
if (div >= 64 && max(board) < 64 && avg > best) best = avg;
}
return best;
}